Learning-based path planning algorithms have emerged as pivotal solutions for robot navigation in complex environments due to their efficient state-space exploration capabilities. However, existing approaches suffer from a critical limitation: sensitivity to map scale variations. In time-critical applications such as post-disaster rescue operations, robots must rapidly compute optimal trajectories on large-scale maps while ensuring precise target localization. To address this challenge, we propose Real Time Neural Informed RRT* (RTNI-RRT*), a novel path planning algorithm that leverages neural guidance for real-time performance. Our approach integrates three key components: (1) a Real Time Neural Concentrate module that processes point cloud data within an elliptical subset of the state space through a PointNet++ network, enabling accurate guided-state inference; (2) a Real Time Neural Combine module that guarantees guided-state connectivity to enhance planning robustness; and (3) a Real Time Extract module that dynamically retrieves optimal paths from the maintained state distribution. We evaluate RTNI-RRT* across diverse environmental configurations. Experimental results demonstrate that our method achieves an average $\mathbf{5 8 \%}$ reduction in terms of iteration count and runtime compared to the NRRT* baseline. Code is available at: RTNI-RRT_star.
Recent advances in diffusion models have significantly enhanced the quality of data synthesis. However, current methods in synthesizing driving-view images neglect the importance of generating diverse weather conditions. To address the long tail effect of weather in autonomous driving data, we propose WeatherDrive, an innovative street view generation framework, changing weather conditions and time in multi-view images freely and stably through text description by introducing a Weather Transfer Module (WTM). Besides, multi-modal inputs, including text prompts, camera poses, 3D bounding boxes and road maps, enable us to achieve more precise 3D geometry control. Furthermore, our approach introduces Weighted Cross-View Attention (WCVA) to keep consistency across multiple camera views. Comprehensive qualitative and quantitative assessments on the nuScenes dataset demonstrate WeatherDrive's superior ability to generate high-quality multi-view driving-scene images across diverse weather conditions and time periods.
Image fusion generates more information-rich integrated images by integrating multi-source image data, and boasts extensive applications in fields such as environmental monitoring and battlefield reconnaissance. Although existing technologies can achieve favorable multi-modal image fusion in non-interfering scenarios and aid in the presentation of scene information, they exhibit significant limitations when confronted with the degradation of low-quality source images induced by extreme scenarios, including low light, haze, rain and snow, motion blur, and strong noise interference. To tackle these challenges, this paper proposes a novel anti-degradation multi-modal image fusion framework guided by supplementary text instructions. Firstly, the Low-rank mechanism is employed to conduct targeted fine-tuning on the CLIP vision-language large model, enabling accurate capture of scene degradation features such as low light, noise, and motion blur, and facilitating the generation of semantic descriptions. Secondly, a heterogeneous mixture-of-experts feature extraction backbone network is designed, which leverages scene degradation semantic descriptions to dynamically regulate heterogeneous expert modules. Meanwhile, a mutual information loss function is introduced to optimize the sparsity of experts during multi-task processing, thereby avoiding redundant computations. Finally, each module undergoes training-testing collaborative optimization to enhance the controllability of the fusion process and improve the generalization capability of the model. Experiments on 4 mainstream benchmark datasets (LLVIP, MSRS, M3FD, ROAD) demonstrate that the proposed text-semantic-guided sparse mixture-of-experts mechanism exhibits significant advantages over state-of-the-art methods in terms of image fusion performance and degradation correction. This research outcome provides new insights for multi-modal image fusion in complex degraded environments and is expected to promote technological advancement in fields such as remote sensing images and medical images.
Owing to the inherent limitation of coarse-grained supervision, contrastive vision-language models such as CLIP typically exhibit suboptimal performance in fine-grained attribute understanding tasks. We propose SubCLIP, a lightweight framework designed to enhance the capability of the CLIP in capturing subtle visual-semantic distinctions. SubCLIP enhances fine-grained attribute understanding by decomposing the positive and negative captions of an image into subtexts containing attribute words and corresponding nouns and introducing a multi-granularity contrastive learning objective. Specifically, we introduce Hierarchical Subtext Augmentation, which leverages a large language model to generate fine-grained, subtexts from existing annotations, thereby providing denser and more informative training signals. To fully exploit these subtexts, we propose Multi-Granularity Contrastive Learning, a unified strategy that jointly optimizes global image-text alignment, intra-sample attribute-level contrast, and subtext-level compositional alignment. This hierarchical design enables SubCLIP to learn both high-level semantic associations and subtle attribute differentiations. On the FG-OVD benchmark, SubCLIP significantly improves fine-grained attribute retrieval performance. When using ROI features for retrieval, it achieves Top-1 accuracy rates of 62.4 % on the hard samples, surpassing previous state-of-the-art methods. Our results demonstrate SubCLIP's effectiveness in enhancing fine-grained vision-language understanding, delivering robust performance while maintaining computational efficiency and domain adaptability. Our code is available at https://github.com/mop134679852/SubCLIP.
Falls pose a serious challenge to the health and safety of the elderly population. In crowded public places, such incidents may also trigger secondary disasters such as stampede, so rapid identification of fall behavior is crucial for maintaining public safety. Traditional fall detection algorithms have significant deficiencies in terms of insufficient feature extraction, single detection method, and poor real-time performance, which make it difficult to cope with complex fall scenarios and real-time detection requirements. In order to overcome the shortcomings of traditional fall detection algorithms, this study improves the PPYOLOE algorithm. We optimize feature extraction using Depthwise Separable Convolution (DSConv) and introduce High-Level Screening-feature Fusion Pyramid Networks (HS-FPN) and Efficient Multi-Scale Attention (EMA) to enhance important features. The improved algorithm significantly improves the detection performance and computational efficiency. Compared with the traditional PPYOLOE algorithm, the fall detection algorithm designed in this paper shows significant advantages in detection accuracy and real-time performance. In the experiment, Our improved algorithm has achieved a 3.5% increase in accuracy on the mAP@0.5 metric, it reaches 94.4% and achieves 24fps on GPU, which realizes the high precision detection of fall events and meets the deployment requirements in low computation scenarios. The real-time and high precision of the algorithm makes it suitable for various monitoring scenarios, which improves the social security level and emergency response capability.
